Personal genome sequences from next generation sequencing technologies are permeating clinical care via diagnostic testing. Clinical labs are under pressure to handle this data in unambiguous, reproducible ways, and provide interpretations that are comparable across testing sites, but face many difficulties to do so. Over the last several years the 'variant file', has emerged as the common currency for exchange and analyses of personal genome sequences for research, and now clinical purposes. These files describe every position in a personal genome that differs from the reference GenBank genome sequence. Given their widespread use, the variant file is a logical starting point for designing a format suitable for clinical applications. Currently variant file styles are widely divergent, ther is not uniformity in the way that complex variants are annotated and genomics has not embraced the use of medical data standards. The diagnostic genomics community, aware of these issues, has mobilized a working group to provide recommendations and requirements to unify variant annotation, to improve public health and clinical applications. For example, without clear standards for capturing variant sequence data, sharing data for the following applications is hindered: across laboratories for quality assurance, with databases for making an interpretation, with a patient record for future use. This proposal addresses the problems with polymorphic variant description by providing novel algorithms to define sequence variants and by developing file formats and software to communicate this information, using guidance from the clinical diagnostic community. The standardized format, VCFclin, and co-developed software tools will end information loss and ambiguity as genomics data flow from sequencing machines, through variant calling and analysis pipelines to interpretation and clinical use.